Covid-19 Cases and Deaths analysis around the world

Shubh Lavti, IIT Gandhinagar, sunil.sl@iitgn.ac.in

Chetan Kishore, IIT Gandhinagar, chetan.k@iitgn.ac.in

Introduction

The Covid-19 pandemic has shook the entire world since it's onset from December 2019. Covid-19 has spread across the globe in less than 4 months and it became very important for WHO to study the growth of virus in different regions. Every region had it's peak at different time and to study the past peaks many have predicted any future wave. Also the fatality rates will help in proper distribution of vaccine.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import plotly.express as px
import plotly
plotly.offline.init_notebook_mode()

df = pd.read_csv("./data/WHO-COVID-19-global-table-data.csv")
df

	Name	WHO Region	Cases - cumulative total	Cases - cumulative total per 100000 population	Cases - newly reported in last 7 days	Cases - newly reported in last 7 days per 100000 population	Cases - newly reported in last 24 hours	Deaths - cumulative total	Deaths - cumulative total per 100000 population	Deaths - newly reported in last 7 days	Deaths - newly reported in last 7 days per 100000 population	Deaths - newly reported in last 24 hours
0	Global	abs	493392853	6329.980092	7968332	102.229659	1157995	6165833	79.104511	23607	0.302866	3930
1	United States of America	Americas	79501007	24018.239000	207355	62.645000	35049	975540	294.723000	3657	1.105000	610
2	India	South-East Asia	43031958	3118.248000	7518	0.545000	1033	521530	37.792000	401	0.029000	43
3	Brazil	Americas	30040129	14132.580000	157732	74.206000	27331	660528	310.750000	1287	0.605000	216
4	France	Europe	25623682	39397.193000	926709	1424.843000	161950	139706	214.802000	755	1.161000	128
...	...	...	...	...	...	...	...	...	...	...	...	...
233	Pitcairn Islands	Western Pacific	0	0.000000	0	0.000000	0	0	0.000000	0	0.000000	0
234	Saint Helena	Africa	0	0.000000	0	0.000000	0	0	0.000000	0	0.000000	0
235	Tokelau	Western Pacific	0	0.000000	0	0.000000	0	0	0.000000	0	0.000000	0
236	Turkmenistan	Europe	0	0.000000	0	0.000000	0	0	0.000000	0	0.000000	0
237	Tuvalu	Western Pacific	0	0.000000	0	0.000000	0	0	0.000000	0	0.000000	0

238 rows × 12 columns

Col = ["Cases - cumulative total per 100000 population","Cases - newly reported in last 7 days per 100000 population","Deaths - cumulative total per 100000 population","Deaths - newly reported in last 7 days per 100000 population"]
# required_columns = ["Name", "WHO Region","Cases - cumulative total", "Cases - newly reported in last 7 days", "Cases - newly reported in last 24 hours","Deaths - cumulative total","Deaths - newly reported in last 7 days", "Deaths - newly reported in last 24 hours"]
covid_df = df.drop(Col, axis=1)
# covid_df = covid_df.sort_values("Cases - cumulative total")
covid_df

	Name	WHO Region	Cases - cumulative total	Cases - newly reported in last 7 days	Cases - newly reported in last 24 hours	Deaths - cumulative total	Deaths - newly reported in last 7 days	Deaths - newly reported in last 24 hours
0	Global	abs	493392853	7968332	1157995	6165833	23607	3930
1	United States of America	Americas	79501007	207355	35049	975540	3657	610
2	India	South-East Asia	43031958	7518	1033	521530	401	43
3	Brazil	Americas	30040129	157732	27331	660528	1287	216
4	France	Europe	25623682	926709	161950	139706	755	128
...	...	...	...	...	...	...	...	...
233	Pitcairn Islands	Western Pacific	0	0	0	0	0	0
234	Saint Helena	Africa	0	0	0	0	0	0
235	Tokelau	Western Pacific	0	0	0	0	0	0
236	Turkmenistan	Europe	0	0	0	0	0	0
237	Tuvalu	Western Pacific	0	0	0	0	0	0

238 rows × 8 columns

new_df = covid_df
uniqueValsList = list(set(df["WHO Region"]))

Continents = {}
for i in uniqueValsList:
    Continents[i] = []
for i in range(len(df)):
    Continents[df["WHO Region"][i]].append(df["Name"][i]) 
del Continents['abs']
del Continents['Other']
listconti = list(Continents.keys())
Countries = []

for i in Continents:
    Countries.append(len(Continents[i]))
Countries
fig = plt.figure(figsize = (20,5))
plt.bar(listconti,Countries,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Countries")
plt.title("Number of Countries in each Region")

myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Countries in each Region")
ax=plt.figure()
ax.set_facecolor('blue')
plt.pie(Countries,labels= listconti,startangle=90,explode = myexplode,shadow=True)
plt.show()

Cases per region

Total Cumulative cases

Cases_per_region = {}
for i in listconti:
    Cases_per_region[i]=0
for i in range(len(covid_df)):
    for j in range(len(listconti)):
        if df["WHO Region"][i] == listconti[j]:
            Cases_per_region[listconti[j]]+=covid_df["Cases - cumulative total"][i]
            break

x=list(Cases_per_region.keys())
y=list(Cases_per_region.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Cases")
# plt.title("Number of Cases in each Region")
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Cases in each Region")
ax=plt.figure()
ax.set_facecolor('blue')
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)
plt.show()

Cases_per_region_7 = {}
for i in listconti:
    Cases_per_region_7[i]=0
for i in range(len(covid_df)):
    for j in range(len(listconti)):
        if df["WHO Region"][i] == listconti[j]:
            Cases_per_region_7[listconti[j]]+=covid_df["Cases - newly reported in last 7 days"][i]
            break

x=list(Cases_per_region_7.keys())
y=list(Cases_per_region_7.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Cases")
# plt.title("Number of Cases in each Region")
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Cases in each Region in last 7 days")
ax=plt.figure()
ax.set_facecolor('blue')
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)

plt.show()

Case_per_region_24 = {}
for i in listconti:
    Case_per_region_24[i]=0
for i in range(len(covid_df)):
    for j in range(len(listconti)):
        if df["WHO Region"][i] == listconti[j]:
            Case_per_region_24[listconti[j]]+=covid_df["Cases - newly reported in last 24 hours"][i]
            break

x=list(Case_per_region_24.keys())
y=list(Case_per_region_24.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Cases")
# plt.title("Number of Cases in each Region")
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Cases in each Region in last 24 hours")
ax=plt.figure()
ax.set_facecolor('blue')
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)
plt.show()

T = {}
for i in Cases_per_region:
    T[i] = [Cases_per_region_7[i],Case_per_region_24[i]]

X = []
y1 = []
y2 = []
cnt = 0
for i in Case_per_region_24:
    
    X.append(cnt+2)
    y1.append(Cases_per_region_7[i])
    y2.append(Case_per_region_24[i])
    cnt+=2
labels = list(Case_per_region_24.keys())

fig = plt.figure(figsize=(20,5))
plt.ylabel("Cases in million")
plt.xlabel("WHO Region")
plt.title("Trend of Covid-19 Cases")
plt.plot(X,y1,label = "Last 7 Days")
plt.plot(X,y2,label = "Last 24 Hours")
plt.xticks(X,labels)
plt.legend()
plt.show()

y3 = []

for i in range(len(y1)):
    y3.append(y2[i]/y1[i])

fig = plt.figure(figsize=(20,5))
plt.plot(X,y3,label = "Ratio of Case in last 24 hours to last 7 days")
plt.xlabel("WHO Region")
plt.ylabel("Ratio")
plt.xticks(X,labels)
plt.legend()
plt.show()
y3

[0.15027388861779462,
 0.16017252249332714,
 0.09321024419297201,
 0.1372273576070679,
 0.144270942744439,
 0.14464966780054914]

From this we can say that the rise in case in last 24 hours is very less in Africa while it has been extremely high for the American and European Countries. Now let us move to the deaths due to Covid-19.

Deaths_perRegion = {}
Deaths_perRegion_7 = {}
Deaths_perRegion_24 = {}
for i in listconti:
    Deaths_perRegion[i]=0
    Deaths_perRegion_7[i]=0
    Deaths_perRegion_24[i]=0
for i in range(len(covid_df)):
    for j in range(len(listconti)):
        if df["WHO Region"][i] == listconti[j]:
            Deaths_perRegion[listconti[j]]+=covid_df["Deaths - cumulative total"][i]
            Deaths_perRegion_7[listconti[j]]+=covid_df["Deaths - newly reported in last 7 days"][i]
            Deaths_perRegion_24[listconti[j]]+=covid_df["Deaths - newly reported in last 24 hours"][i]
            break

x=list(Deaths_perRegion.keys())
y=list(Deaths_perRegion.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Deaths")
plt.title("Number of Deaths in each Region")
plt.show()
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Deaths in each Region")
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)
fig.set_facecolor('blue')
plt.show()


x=list(Deaths_perRegion_7.keys())
y=list(Deaths_perRegion_7.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Deaths")
plt.title("Number of Deaths in each Region in last 7 days")
plt.show()
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Deaths in each Region in last 7 days")
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)
fig.set_facecolor('blue')
plt.show()


x=list(Deaths_perRegion_24.keys())
y=list(Deaths_perRegion_24.values())


fig = plt.figure(figsize = (20,5))
plt.bar(x,y,color='blue',width = 0.2)
plt.xlabel("WHO Region")
plt.ylabel("Deaths")
plt.title("Number of Deaths in each Region in last 24 hours")
plt.show()
myexplode = [0.1,0.1,0.1,0.1,0.1,0.1]
plt.title("Number of Deaths in each Region in last 24 hours")
plt.pie(y,labels= x,startangle=90,explode = myexplode,shadow=True)
fig.set_facecolor('blue')
plt.show()

x = []
y1 = []
y2 = []
y3 = []
cnt = 0
for i in listconti:
    x.append(cnt+2)
    y1.append(Case_per_region_24[i])
    y2.append(Deaths_perRegion_24[i])
    a = 100*Deaths_perRegion_24[i]/Case_per_region_24[i]
    y3.append(a)
    cnt+=2

fig = plt.figure(figsize = (20,5))
plt.plot(x,y3)
plt.xlabel("WHO Region")
plt.ylabel("Fatality Rate in percent")
plt.title("Deaths and Cases per Region")
plt.xticks(x,labels)
plt.show()

covid_df = covid_df.sort_values("Cases - cumulative total")
top_five = covid_df[len(covid_df)-6:len(covid_df)-1]
top_five = top_five.sort_values("Cases - cumulative total")
top_five

	Name	WHO Region	Cases - cumulative total	Cases - newly reported in last 7 days	Cases - newly reported in last 24 hours	Deaths - cumulative total	Deaths - newly reported in last 7 days	Deaths - newly reported in last 24 hours
5	Germany	Europe	22265788	1161279	201729	131036	1645	328
4	France	Europe	25623682	926709	161950	139706	755	128
3	Brazil	Americas	30040129	157732	27331	660528	1287	216
2	India	South-East Asia	43031958	7518	1033	521530	401	43
1	United States of America	Americas	79501007	207355	35049	975540	3657	610

top_five_country = list(top_five["Name"])
top_five_country

['Germany', 'France', 'Brazil', 'India', 'United States of America']

Although Europe has very high number of cases in Recent times but the fatality rate is still pretty less as compared to other WHO Regions. The Fatality rate is highest in American Continent. Thus the distribution of vaccine can be proposed on the basis of this rates.

date_df = pd.read_csv("./data/WHO-COVID-19-global-data.csv")
date_df

	Date_reported	Country_code	Country	WHO_region	New_cases	Cumulative_cases	New_deaths	Cumulative_deaths
0	2020-01-03	AF	Afghanistan	EMRO	0	0	0	0
1	2020-01-04	AF	Afghanistan	EMRO	0	0	0	0
2	2020-01-05	AF	Afghanistan	EMRO	0	0	0	0
3	2020-01-06	AF	Afghanistan	EMRO	0	0	0	0
4	2020-01-07	AF	Afghanistan	EMRO	0	0	0	0
...	...	...	...	...	...	...	...	...
195757	2022-04-03	ZW	Zimbabwe	AFRO	195	246481	2	5446
195758	2022-04-04	ZW	Zimbabwe	AFRO	44	246525	0	5446
195759	2022-04-05	ZW	Zimbabwe	AFRO	87	246612	5	5451
195760	2022-04-06	ZW	Zimbabwe	AFRO	132	246744	0	5451
195761	2022-04-07	ZW	Zimbabwe	AFRO	59	246803	2	5453

195762 rows × 8 columns

Country = {}
dates = []
for i in range(len(date_df)):
    if date_df["Country"][i] in Country:
        Country[date_df["Country"][i]].append(date_df["New_cases"][i])
    else:
        Country[date_df["Country"][i]] = [date_df["New_cases"][i]]

    if date_df["Date_reported"][i] not in dates:
        dates.append(date_df["Date_reported"][i])
# len(dates)
# Country

Months = ["January","February","March","April","May","June","July","August","September","October","November","December"]

for l in top_five_country:
    DATE = []
    x=[]
    cnt = 0
    y=[]
    for i in range(len(dates)//30):
        DATE.append(Months[int(i%12)])
        y.append(0)
    for i in range(len(list(Country[l]))):
        x.append(cnt+1)
        cnt+=1


    fig = plt.figure(figsize = (20,5))
    plt.plot(x,list(Country[l]),color="black")
    xd = []
    flag = 20
    for i in range(7):
        xd.append(flag)
        flag+=120

    DATE = ["January 23, 2020","May 23,2020","September 20,2020","January 18,2020","May 20,2021","September 17,2021","January 15,2022"]
    plt.xticks(xd,DATE)
    plt.title("Covid Cases in "+str(l))
    plt.xlabel("Timeline")
    plt.ylabel("New Cases per Day")
    plt.show()

Country_d = {}
dates = []
for i in range(len(date_df)):
    if date_df["Country"][i] in Country_d:
        Country_d[date_df["Country"][i]].append(date_df["New_deaths"][i])
    else:
        Country_d[date_df["Country"][i]] = [date_df["New_deaths"][i]]

    if date_df["Date_reported"][i] not in dates:
        dates.append(date_df["Date_reported"][i])
len(dates)
Months = ["January","February","March","April","May","June","July","August","September","October","November","December"]

for l in top_five_country:
    DATE = []
    x=[]
    cnt = 0
    y=[]
    for i in range(len(dates)//30):
        DATE.append(Months[int(i%12)])
        y.append(0)
    for i in range(len(list(Country_d[l]))):
        x.append(cnt+1)
        cnt+=1


    fig = plt.figure(figsize = (20,5))
    plt.plot(x,list(Country_d[l]),color="black")
    xd = []
    flag = 20
    for i in range(7):
        xd.append(flag)
        flag+=120

    DATE = ["January 23, 2020","May 23,2020","September 20,2020","January 18,2020","May 20,2021","September 17,2021","January 15,2022"]
    plt.xticks(xd,DATE)
    plt.title("New Deaths in "+str(l))
    plt.xlabel("Timeline")
    plt.ylabel("New Deaths")
    plt.show()

As we can see from the above graphs that the peak in covid-19 cases and deaths are distinct for all the five countries. In India we can observe three waves (September 2020, May 2021, January 2022) in which the most devastating wave was the second wave.